Little is known about the neurocircuit and neurobiological bases for regulation of feeding and metabolism, and this has greatly limited progress in understanding and treating obesity and feeding disorders. Lack of knowledge in this area is due, in large part, to complexity within brain regions controlling these processes - namely those that lie within and are connected to the hypothalamus. Each anatomic subregion contains many different types of neurons, each controlling unrelated, opposite or unknown functions. While general information exists regarding connectivity between subregions, this provides little mechanistic insight because the functions of the different neurons within each subregion are complex and/or unknown, and the labeled lines connecting specific upstream neurons to specific downstream neurons are also not known. In essence, we lack a wiring diagram for hypothalamic control of behavior and physiology. With recent technological advances, enabled by neuron-specific Cre-expressing mice, it is now possible in a cell-specific fashion to establish connectivity and function. The present proposal utilizes such approaches to delineate the neurocircuitry underlying leptin regulation of energy balance. These studies build upon our recent discovery that the majority of leptin's anti-obesity effects are mediated by leptin receptors on GABAergic neurons. In Aim 1, we set out to identify the source of leptin-responsive GABAergic input to POMC and AgRP neurons. In preliminary studies, using Cre-dependent monosynaptic rabies mapping and channelrhodopsin-assisted circuit mapping, we have determined that, for POMC neurons, leptin-responsive GABAergic input is entirely from local neurons (all within the arcuate), while for AgRP neurons, very strong leptin-responsive GABAergic input comes from the dorsomedial hypothalamus (DMH). Given the key functional importance of these DMH afferents to AgRP neurons in controlling hunger (as shown in Aim 2), we are using Single-Neuron RNA-Seq to determine their identity, to detect genes likely to suggest function (such as those involved in neurotransmitter synthesis and transport, neuropeptides and receptors), and to begin a search for possible drug targets. In Aim 2, we are using optogenetic techniques in awake, behaving mice to determine the role of DMH leptin receptor-expressing GABAergic inputs to AgRP neurons in regulating hunger. In preliminary studies we have discovered that optogenetic activation of these afferents completely blocks hunger, even that caused by fasting. Thus, these neurons are previously unknown potent regulators of hunger. Finally, in Aim 3, we are performing optetrode recordings in awake, behaving mice to determine, in real time, the firing rate of AgRP neurons and their DMH leptin-responsive GABAergic afferents. By using light/ChR2- evoked spiking to identify neurons, these studies address previously inaccessible questions regarding effects of behavioral and physiologic perturbations on in vivo firing rates. In total, these studies should significantly advance our understanding of how leptin, and the circuits it regulates, control eating and energy balance.